A recently developed rehabilitation program for subacute-chronic post-stroke hemiparesis, Constraint-Induced Movement therapy (CI therapy), has become increasingly adopted, and has been investigated worldwide in over 300 studies to date, due to its successful translation from basic neuroscience research studies and its demonstrated efficacy in numerous randomized controlled clinical trials. The therapy's principal value is for improving the spontaneous use of the paretic upper extremity in the real-life setting when the limb has sufficient movement ability to complete daily living tasks. The reduced use of the paretic limb post-stroke, termed learned nonuse (LNU), is hypothesized to result from the behavioral suppression of the paretic limb's participation in real-world activities due to its repeated ineffectual movement attempts and simultaneously effective compensation by the less-affected upper extremity. However, although basic neuroscience research has supported the LNU formulation in animal models of hemiparesis through detailed observations, and outcomes from numerous CI therapy trials in the chronic phase have suggested that stroke patients had developed LNU before treatment, the LNU hypothesis thus far has not been tested in humans. The proposed 2 year research will address this knowledge gap by longitudinally evaluating acutely hemiparetic stroke patients (N = 66) on validated measures of spontaneous paretic limb use in the life setting and maximal movement ability in the laboratory. Patients will be evaluated at 7-10 days post-stroke and again at 3 months. Nonuse (defined as the difference between maximal movement ability and simultaneous spontaneous limb use) is hypothesized to increase during this period, consistent with a learning mechanism. The proposed research will prospectively determine whether nonuse significantly increases over this time interval, as predicted by the LNU hypothesis. Concurrent clinical evaluations of coordination, dexterity, tactile sensitivity, grip strength, limb tone, pain, depression, spatial attention, and general stroke impairment will determine whether these factors modulate LNU. The extent of acute ischemic injury on Diffusion-Weighted Imaging on brain MRI, and in particular the extent of injury to the cerebral course of the corticospinal tract, will also be evaluated for their contribution to LNU. The findings will determine whether evidence for LNU occurs in the acute-subacute poststroke phase and whether it has clinical or structural neuroimaging correlates. The findings will help to improve understanding the sequential longitudinal changes of neurologic functions during acute-subacute hemiparetic stroke that relate to movement of the paretic limb and how these may modulate limb nonuse. The findings will be used to support subsequent extended research proposals on the epidemiology and neuroscientific mechanisms of LNU after stroke, which will be important for improving the delivery of neurologic rehabilitation to stroke survivors. PUBLIC HEALTH RELEVANCE: This proposal will test hypotheses concerning the origin of learned nonuse, which is the reduction in the amount of daily use by the arm that has been impaired by stroke, when the arm nonetheless has sufficient movement ability to perform daily living activities. Nonuse appears to commonly occur after stroke and has been found to significantly and enduringly improve after specific forms of movement therapy, particularly Constraint-Induced Movement therapy (CI therapy). The research is expected to provide pilot data to help determine how frequently learned nonuse occurs in stroke and what clinical factors predict its onset, which in turn would help to improve models for stroke rehabilitation that will be tested further in later, more comprehensive funding proposals.